The Role Of Deregulated HMGA2 Expression With Promoter Methylation Of p16 In Myeloproliferative Neoplasms

Myeloproliferative Neoplasms (MPNs) are characterized by clonal proliferative hematopoiesis with increased mature blood cells. The signal-activating mutations such as JAK2V617F increase blood cells, but it remains uncertain how an abnormal hematopoietic cell clone expands in MPNs. We have recently showed that overexpression of the high mobility group AT-hook 2 (HMGA2) causes proliferative hematopoiesis with providing a clonal growth advantage to hematopoietic cells in mice (Ikeda et al, Blood, 2011), suggesting the possibility that HMGA2 contributes to the pathogenesis of MPNs. However, since only a few studies have evaluated expression of HMGA2 mRNA in patients with MPNs, the role of HMGA2 in the pathogenesis of MPNs is yet unclear. MPNs also show mutations in epigenetic modifiers involving DNA methylation such as polycomb group genes (PcG) and aberrant expressions of micro RNAs (miRNA) that negatively regulate expressions of targeted genes. Interestingly, deficiency in either PcG-related BMI1 (Oguro et al, J Exp Med, 2012) or let-7-family miRNA (Mayr et al, Science, 2007) causes deregulation of HMGA2 expression, leading to its oncogenic activity in part by negatively regulating tumor suppressor p16. Thus, in this study, to clarify the role of HMGA2 in MPNs, we investigated expression of HMGA2 mRNA in peripheral granulocytes of 56 patients with MPNs including 23 polycythemia vera (PV), 26 essential thrombocythemia (ET) and 7 primary myelofibrosis (PMF) along with clinical findings, JAK2V617F allele burden, expressions of BMI1 mRNA and let-7-family miRNAs, and promoter methylation of p16. Quantitative RT-PCR (qPCR) showed significantly higher expression of HMGA2 mRNA relative to internal control HPRT1 mRNA in PMF (mean ± SD; 31.7 ± 42.8, p<0.01), but not PV (15.7 ± 53.2) or ET (2.14 ± 7.70), compared with 12 healthy volunteers (HV; 0.431 ± 0.366). In addition, deregulated HMGA2 expression (>1.2), which was determined as relative expression level above mean + 2SD of HMGA2 mRNA in 12 HV, was most frequently detected in patients with PMF [7/7 (100%)] (p<0.01), compared with PV [5/23 (21.7%)] and ET [6/26 (23.1%)]. We also found a significant positive correlation in expression levels of HMGA2 mRNA with serum LDH values (r=0.531, p<0.01) rather than JAK2V617F allele burden (r=0.25, p=0.08). These data suggested that expression of HMGA2 mRNA independently correlated with disease phenotype and status in MPNs. We next explored the cause of deregulated expression of HMGA2 mRNA and found lower expression of let-7a (0.19 ± 0.13 vs. 0.42 ± 0.39, p=0.04) and -7c (0.57 ± 0.60 vs. 1.14 ± 0.94, p=0.06) rather than -7b (p=0.2) by qPCR, in patients with deregulated expression of HMGA2 mRNA compared with other patients. However, HMGA2-involved chromosomal abnormality in 12q13-15 was not detected in any patient, and there was no difference in expression of BMI1 mRNA between patients with deregulated expression of HMGA2 mRNA and other patients. Thus, decreased expression of let-7 miRNAs might contribute to deregulated expression of HMGA2 mRNA in MPNs. Finally, we investigated correlation of deregulated expression of HMGA2 mRNA with promoter methylation of p16. Methylation-specific PCR assay detected promoter methylation of p16 in 17/56 (30.4%) patients with MPNs. Strikingly, patients with deregulated expression of HMGA2 mRNA significantly more often showed promoter methylation of p16 compared with other patients [10/18 (55.6%) vs. 7/38 (18.4%), p<0.01]. Furthermore, patients with promoter methylation of p16 showed higher expression levels of HMGA2 mRNA than patients without the methylation, especially in patients with PMF (2.33 ± 0.90 vs. 70.9 ± 38.3, p=0.01). In conclusion, deregulated expression of HMGA2 in association with decreased expression of let-7 miRNAs may play a crucial role in the pathogenesis of MPNs possibly through p16.